Module: Regexp::Lexer

Defined in:
lib/regexp_parser/lexer.rb

Overview

A very thin wrapper around the scanner that breaks quantified literal runs, collects emitted tokens into an array, calculates their nesting depth, and normalizes tokens for the parser, and checks if they are implemented by the given syntax flavor.

Constant Summary collapse

OPENING_TOKENS = 
[:capture, :options, :passive, :atomic, :named,
 :lookahead, :nlookahead, :lookbehind, :nlookbehind
].freeze
CLOSING_TOKENS = 
[:close].freeze

Class Method Summary collapse

Class Method Details

.ascend(type, token) ⇒ Object 



51
52
53
54
55
56
57
58
59
 
# File 'lib/regexp_parser/lexer.rb', line 51

def self.ascend(type, token)
  if type == :group or type == :assertion
    @nesting -= 1 if CLOSING_TOKENS.include?(token)
  end

  if type == :set or type == :subset
    @set_nesting -= 1 if token == :close
  end
end

.break_literal(token) ⇒ Object

called by scan to break a literal run that is longer than one character into two separate tokens when it is followed by a quantifier



73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
 
# File 'lib/regexp_parser/lexer.rb', line 73

def self.break_literal(token)
  text = token.text
  if text.scan(/./mu).length > 1
    lead = text.sub(/.\z/mu, "")
    last = text[/.\z/mu] || ''

    if RUBY_VERSION >= '1.9'
      lead_length = lead.bytesize
      last_length = last.bytesize
    else
      lead_length = lead.length
      last_length = last.length
    end

    @tokens.pop
    @tokens << Regexp::Token.new(:literal, :literal, lead, token.ts,
                                 (token.te - last_length), @nesting, @set_nesting)

    @tokens << Regexp::Token.new(:literal, :literal, last,
                                 (token.ts + lead_length),
                                 token.te, @nesting, @set_nesting)
  end
end

.descend(type, token) ⇒ Object 



61
62
63
64
65
66
67
68
69
 
# File 'lib/regexp_parser/lexer.rb', line 61

def self.descend(type, token)
  if type == :group or type == :assertion
    @nesting += 1 if OPENING_TOKENS.include?(token)
  end

  if type == :set or type == :subset
    @set_nesting += 1 if token == :open
  end
end

.merge_literal(current) ⇒ Object

called by scan to merge two consecutive literals. this happens when tokens get normalized (as in the case of posix/bre) and end up becoming literals.



99
100
101
102
103
 
# File 'lib/regexp_parser/lexer.rb', line 99

def self.merge_literal(current)
  last = @tokens.pop
  replace = Regexp::Token.new(:literal, :literal, last.text + current.text,
                                 last.ts, current.te, @nesting, @set_nesting)
end

.scan(input, syntax = "ruby/#{RUBY_VERSION}", &block) ⇒ Object 



13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
 
# File 'lib/regexp_parser/lexer.rb', line 13

def self.scan(input, syntax = "ruby/#{RUBY_VERSION}", &block)
  syntax = Regexp::Syntax.new(syntax)

  @tokens = []
  @nesting, @set_nesting = 0, 0

  last = nil
  Regexp::Scanner.scan(input) do |type, token, text, ts, te|
    type, token = *syntax.normalize(type, token)
    syntax.check! type, token

    ascend(type, token)

    break_literal(last) if type == :quantifier and
      last and last.type == :literal

    current = Regexp::Token.new(type, token, text, ts, te,
                                @nesting, @set_nesting)

    current = merge_literal(current) if type == :literal and
      last and last.type == :literal

    last.next(current) if last
    current.previous(last) if last

    @tokens << current
    last = current

    descend(type, token)
  end

  if block_given?
    @tokens.each {|t| block.call(t)}
  else
    @tokens
  end
end